Same as it was done for Emozionalmente, where we crowdsourced emotional speech data using a gamified web application, here we want to collect emotional conversational data obtained during 1vs1 and many-to-many conversations (multi-agent, multi-user setting)